- Email: [email protected]
Pelvic nerve grafts restore bladder function in denervated rats
Pelvic nerve grafts restore bladder function in denervated rats Toru Kono, MD, PhD, Masanobu Miyata, MD, PhD, Yasuhiro Yamamoto, MD, PhD, Akitoshi Kakisaka, MD, PhD, Sunao Yachiku, MD, PhD, and Shinichi Kasai, MD, PhD, Hokkaido, Japan
Background. Autonomic nerve preservation techniques for use during surgery for rectal cancer have improved. Nevertheless, in some patients pelvic nerves must be sacrificed to excise all tumor. For these patients, nerve reconstruction at the time of operation by using nerve grafting would be useful. A rat model of this type of nerve reconstruction is described. Methods. Animals were divided into three groups. In the sham control group, a pelvic exploration was conducted without division of the pelvic nerves. In the nerve ablation group, both pelvic nerves were excised segmentally. In the graft group, both pelvic nerves were excised and genitofemoral nerves were interposed bilaterally. At 2-week intervals postoperatively, animals from each group underwent cystometry under urethane anesthesia and neuronal tracing using fragment C of tetanus toxin for demonstration of axonal transport via regenerated nerves. Results. At 6 weeks postoperatively, 60% of the grafted animals produced rhythmic contractions of the bladder. In neuronal tracing studies at weeks 4 and 6, respectively, 40% and 100% of the nerve-grafted rats had labeled neurons in the sacral parasympathetic nucleus. Conclusions. These findings suggest that pelvic nerve grafting in rats can successfully restore bladder function after surgical injury. Clinical use of pelvic nerve grafting may be indicated for patients whose pelvic nerves must be sacrificed to excise all tumor.(Surgery 1998;123:672-8.) From the Second Department of Surgery and the Department of Urology, Asahikawa Medical College, Hokkaido, Japan
SURGICAL TREATMENT FOR RECTAL CANCER has profound effects on bladder function resulting from autonomic denervation.1,2 Recent studies have precisely localized the nerves that control bladder function.3-5 With this knowledge it has been possible to preserve the autonomic nerves controlling bladder function in many patients undergoing radical resection for rectal cancer. However, in some patients with advanced rectal cancer, autonomic nerve preservation cannot be performed if adequate surgical margins are to be obtained. At the present time, no methods are available for restoring bladder function in these patients. Therefore, methods to restore neural control of bladder function without compromising rectal cancer control must be devised. The feasibility of nerve regeneration in denervated tissues of the peripheral nervous system has been carefully studied.6-8 In addiSupported by research grant (C)(2)07671350 from the Japanese Ministry of Education, Science, and Culture. Accepted for publication Nov. 5, 1997. Reprint requests: Toru Kono, MD, PhD, The Second Department of Surgery, Asahikawa Medical College, 4-5 Nishikagura, Asahikawa, Hokkaido 078, Japan. Copyright © 1998 by Mosby, Inc. 0039-6060/98/$5.00 + 0 11/56/87454
672 SURGERY
tion, nerve grafting is a well-established technique in plastic surgery for repair of damaged nerves.9-16 However, the use of such techniques for autonomic nerves controlling bladder function has not been studied. Methods enabling direction of regenerating autonomic nerve would be valuable for patients who require intentional surgical pelvic nerve denervation to ensure clear surgical margins. We demonstrate here that nerve grafting with transposed genitofemoral nerve can restore bladder function in rats that had previously undergone surgical nerve ablation. This study demonstrates the feasibility of restoration of bladder function after advanced cancer surgery by interpositional nerve grafting of the pelvic nerves. MATERIAL AND METHODS Forty-five Wistar rats were divided into three groups of 15 animals each: (1) sham controls, (2) pelvic nerve ablation, and (3) pelvic nerve grafting. Postoperatively, each 2 weeks for 6 weeks five animals from each group underwent postoperative bladder function tests17 and examination for restoration of axonal transport between the regenerated sciatic nerve and the spinal cord, principally its lumbosacral portion, using fragment C of tetanus toxin (TTC) retrograde tracing methods.18,19
Surgery Volume 123, Number 6
Kono et al. 673
Fig. 1. Scheme illustrates surgical procedures in rats. (A) The pelvic ganglion receives input from the pelvic nerve, which is the final preganglionic parasympathetic pathway controlling bladder function in the rat. (B) In the nerve ablation group, a 5 mm segment was excised from both pelvic nerves. (C) In the graft group, a 5 mm segment was excised from both pelvic nerves and replaced with interpositional grafts of genitofemoral nerves bilaterally. A nerve graft sewn in place restored continuity of the pelvic nerve sheath.
Surgical procedures. Seventy-day-old Wistar rats (250 to 270 gm) were used. All animals were anesthetized with intraperitoneal sodium pentobarbital (25 mg/kg) and underwent surgical exploration to expose the innervation of the pelvic viscera and lower genitourinary tract.20 A midline incision was made from umbilicus to pubis. With the aid of a Nikon dissecting microscope (magnification ×8), the hypogastric nerve, pelvic nerve, major pelvic ganglion, and the nerve fibers to the lower genitourinary tract were identified20 (Fig. 1). The pelvic nerves were identified and exposed bilaterally. In the sham control group, the pelvic nerves were identified but not divided at the time of operation (Fig. 1, A). In the pelvic nerve ablation group, a 5 mm segment of pelvic nerve just distal to the major pelvic ganglion was surgically excised bilaterally (Fig. 1, B). Care was taken to excise all branches of the pelvic nerve trunk, because unsuccessful denervation has been reported when the lateral branches have been spared. In the pelvic nerve graft group, a 5 mm segment of pelvic nerve and all accessory branches were excised bilaterally, and an isograft of genitofemoral nerve was interposed bilaterally (Fig. 1, C). The direction of the graft was reversed so that the bifurcating distal end of the genitofemoral nerve (dividing into its genital and femoral branches) was placed proximally. This method of positioning allows regenerating nerve fibers to converge distally.16 An epineural
anastomosis (proximal and distal) was then performed with 10-0 nylon sutures using microsurgical technique. Bladder function tests. Cystometric bladder function studies in the urethane-anesthetized rat have been well described.17,21,22 Every 2 weeks postoperatively for 6 weeks, five animals from each of the three experimental groups underwent bladder function tests. All bladder function tests were performed by transvesical recording of intravesical pressure. Rats were anesthetized with urethane (1.5 gm/kg, subcutaneously). Through a midline incision of the abdomen, the urinary bladder was exposed and a Teflon catheter (Angiocath, 18gauge; Deseot Medical Inc.) was implanted into the bladder through the dome and secured in place by means of a silk ligature. The bladder catheter was connected via a T-tube to a pressure transducer and a microinjection pump. The urethra was ligated between the prostate and the bladder using a silk thread suture. Warm saline-soaked cotton wool swabs were laid around the exteriorized organ to maintain its temperature and keep it moist. After a 15-minute equilibration period at zero volume, the bladder was rapidly filled with 0.5 ml of warm saline (37° C) to increase and maintain intravesical pressure. This maneuver produced neurogenic rhythmic contractions. Intravesical pressure was recorded continuously on a San-ei polygraph (Model 363; Nippon Koden San-ei) and displayed on a poly-
674 Kono et al.
Fig. 2. A representative tracing illustrates results of bladder function testing for a sham-operated rat (Sham) and a rat that had undergone bilateral removal of the pelvic nerves (Bilateral Denervation). IVP, intravesical pressure.
graph. At the beginning of cystometric measurements, the bladder was emptied through the bladder catheter. In the pelvic nerve graft group, cystometry was performed before and after the grafted pelvic nerve was removed. The rhythmic contractions required 5 to 10 minutes to reach steady state after saline loading. The frequency, duration, and amplitude of rhythmic contractions were determined at this time to evaluate the effects of each preparation. The number, width, and height of contractions were determined during a 4-minute observation period. TTC tracing studies. After bladder function tests, animals underwent TTC tracing tests. TTC (10 µl) was slowly injected into the ventral bladder detrusor muscle using a Hamilton microliter syringe fitted with a 27-gauge needle, and the incision was closed. The TTC (Behring DiagnosticsCalbiochem; molecular weight, 47,000) had been concentrated to approximately 15% (w/v) by ultrafiltration with a Centricon tube (Amicon; 30,000 molecular weight cutoff membrane). All injections were made into animals anesthetized with urethane. Optimal survival times for transsynaptic labeling were determined for normal rats in which TTC had been injected into the wall of the bladder. Transsynaptic labeling was well developed by 3 days
Surgery June 1998 after injection. Therefore, after a transport time of 72 hours for TTC, each animal was deeply anesthetized. Injection of 1000 units of heparin preceded transcardial perfusion with 6% dextran followed by 4% paraformaldehyde in 0.1 mol/L phosphate buffer (PB) at pH 7.4. After perfusion, the lower spinal cord was removed, postfixed in buffered 4% paraformaldehyde at 4° C for 3 hours, and placed in 30% sucrose in PB at 4° C for 24 hours. The spinal cord was subsequently frozen and cut frontally in serial order at 10 µm thickness in a cryostat, serially mounted on gelatinized slides, and heat-dried on a slide warmer at 40° to 50° C. Immunohistochemistry for TTC. TTC was localized immunohistochemically. The sections were treated with endogenous peroxidase (0.2% H2O2 in distilled water for 10 minutes) and incubated overnight at 4° C in a monoclonal antibody that had been raised in mouse against TTC. The antibody (ascites fluid) was diluted 1:500 in 0.3% Triton-X and PB (TX/PB). The sections were washed in PB, incubated at room temperature for 2 hours in biotinylated anti-mouse immunoglobulin G diluted 1:200 in TX/PB, washed again in PB, and then incubated for 1 hour in avidin-biotinhorseradish peroxidase complex (1:50 in TX/PB) using standard procedures and reagents from Vector Laboratories. In all cases, a 5-minute incubation in 3% cobalt chloride in PB was included to intensify the reaction. The tissue was then reacted with diaminobenzidine (55 mg/1,000 ml PB) with 0.03%H2O2. Sections were dried and incubated briefly in a 0.05% osmium tetroxide solution before dehydration and coverslipping. In control experiments, tissue unexposed to TTC was processed normally, and tissue containing TTC was processed without the primary antibody. Both sets of control experiments yielded no labeling. Statistical analyses. Results are given as mean values ± standard error of the mean. Statistical analysis was performed using Student’s t test for paired or unpaired data as appropriate. Findings were considered significant when p < 0.05. RESULTS Bladder function tests. At 2-week intervals postoperatively, five animals from each group underwent bladder function testing (Table I). Rapid distention of the urinary bladder by infusion of saline in sham animals activated a series of high-amplitude rhythmic contractions that were eliminated by bilateral section of the pelvic nerves (Fig. 2). At 2 and 4 weeks, distention of the urinary bladder by infusion of saline in the grafted animals failed to elicit neurogenic rhythmic contractions
Kono et al. 675
Surgery Volume 123, Number 6
Table I. Percentages of animals producing distention-induced rhythmic contractions of the bladder Treatment groups Sham Ablation Graft
2 wk
4 wk
6 wk
60 (3/5) 0 (0/5) 0 (0/5)
40 (2/5) 0 (0/5) 0 (0/5)
60 (3/5) 0 (0/5) 60 (3/5)
Sham, control group (sham procedure); ablation, animals with bilateral excision pelvic nerves; graft, animals with pelvic nerve graft.
Table II. Comparison of amplitude and duration of distention-induced rhythmic bladder contraction at week 6 postoperatively Treatment groups Sham Graft
Frequency (count/min)
Amplitude (cm H2O)
1.3 ± 0.1* 0.7 ± 0.1*
37.0 ± 0.2† 26.6 ± 0.7†
Duration (sec) 27.7 ± 0.2‡ 25.4 ± 0.7‡
*p < 0.05. †p
< 0.05.
‡No
significant difference.
Table III. Percentages of animals with TTC-labeled neurons in the spinal cord Treatment groups Sham Ablation Graft
2 wk
4 wk
6 wk
100 (5/5) 0 (0/5) 0 (0/5)
100 (5/5) 0 (0/5) 40 (2/5)
100 (5/5) 0 (0/5) 100 (5/5)
Sham, control group (sham procedure); ablation, animals with bilateral excision pelvic nerves; graft, animals with pelvic nerve graft.
(Table I). At 6 weeks, 60% of the grafted animals produced distention-induced rhythmic contractions of the bladder (Table I). The amplitude and frequency of these contractions were significantly (p < 0.05) lower than those in control animals, but the duration was not (Table II). These contractions in the grafted animals disappeared after bilateral section of the grafted pelvic nerves (Fig. 3). None of the animals in the ablation group produced high-amplitude rhythmic contractions during the 6-week period of observation (Table I). These findings indicated that at 6 weeks the grafted animals had approached the sham-operated group in their ability to produce bladder contractions, whereas animals in the nerve ablation group were still unable to produce such contractions. TTC tracing studies. Results are shown in Table III and Fig. 4. At 2 weeks, no labeled cells were found in the spinal cord in either the ablation or graft group. In the sham group, many preganglionic neurons labeled by TTC were found in the sacral parasympathetic nucleus (SPN) in the L6-S1 spinal cord; the majority of labeled cells were located in S1. In transverse sections, the TTC-labeled cells were most frequently observed in tight clusters in the lateral intermediate gray (Fig. 4). At week 4, 40% of the grafted animals had labeled neurons in
the SPN, but the number of labeled neurons was much lower than in control animals. At week 6, labeled neurons were observed in the SPN in 100% of the grafted animals, and the number of labeled neurons was greater than at week 4. No staining of lumbosacral parasympathetic neurons was observed in the nerve-ablated animals. These findings demonstrated that regrowth of nerve fibers along the pelvic nerve graft had resulted in regeneration of the distal pelvic nerves during the 4- to 6week postoperative period. DISCUSSION Clinical interest in bladder function after surgical resection (e.g., abdominoperineal resection, low anterior resection) for advanced rectal cancer has resulted from dissatisfaction with what had been accepted as the necessary loss of such function after this operation. It is now possible to identify the location of the pelvic nerves intraoperatively and, in certain patients, to avoid injuring them, thus preserving autonomic nerve control of bladder function. In some patients, however, it is necessary to excise both pelvic nerves to obtain clear surgical margins. Unfortunately, the chance of preserving natural bladder function in these patients has been very small. Patients undergoing
676 Kono et al.
Fig. 3. A representative tracing illustrates results of bladder function testing at week 6 for a rat with pelvic nerve graft (Graft) and a rat that had undergone bilateral removal of the grafted nerves (Bilateral Denervation). IVP, intravesical pressure.
partial excision of the pelvic nerves for malignant disease also face a serious risk of urinary dysfunction. No work has focused on procedures for restoring natural bladder function without compromising focal control of rectal cancer and elimination of recurrence in such patients. It has been demonstrated in animal models that autonomic peripheral nerves can regenerate and reestablish terminal contacts. The use of direct anastomosis or autograft in peripheral nerve repair leads to functional recovery. Nerve grafting is a well-established technique in plastic surgery for repair of damaged motor and sensory nerves9,10,13,14,16 A nerve graft provides a conduit down which regenerating nerve fibers will be directed to meet with the distal end of the transected nerve. Wallerian degeneration in the distal nerve segment (as well as the graft segment itself) occurs once the nerve loses continuity with its proximal segment and cell body. This leaves only the nerve sheath intact. Regenerating fibers may then grow down this sheath directed toward the distal nerve while protected from fibrosis and neuroma formation.13 Degenerating nerve fibers in the distal nerve and graft also may provide nerve growth factor, which will both stimulate nerve fiber growth and direct it toward the distal end of the
Surgery June 1998 nerve.23,24 Nerve graft techniques used by other investigators in rats have demonstrated that cavernous nerve grafting in rats can restore potency after surgical injury.16 To our knowledge, no studies of restoration of bladder function using nerve grafts have been reported. Venous grafts have been shown to facilitate autonomic nerve regeneration.25 In recent years, many biodurable and biodegradable nerve guide conduits manufactured from silicone, polyvinylidone fluoride, propylene, polyglycolic acid, or collagen have been tested.9,11,12,15,26-28 Another approach to improving the success of nerve repair procedures is the search for trophic factors.10,29-31 These factors promote long nerve gap regeneration better than grafting alone. The next step in our work will be to find nerve guide conduits and trophic factors appropriate for restoration of bladder function that can be used in place of nerve grafting. TTC is a nontoxic molecule that undergoes robust retrograde transneuronal transport from the periphery and can be localized using a monoclonal antibody.18,19 The presence of labeling in SPN, one of the portions of the central nervous system controlling detrusor contraction,32 demonstrates that transneuronal transport via regenerated grafted nerve occurs. After injection into the detrusor muscle of the bladder, TTC is taken up by the axon terminals of postganglionic neurons in the major pelvic ganglion and transported in a retrograde fashion to their somas. TTC then moves across synapses to presynaptic elements and passes the pelvic grafted nerve in a retrograde fashion back to the somas of preganglionic neurons in the SPN in the spinal cord. Therefore, retrograde transneuronal transport from the bladder detrusor muscle to the SPN can only be explained by the presence of neuroanatomic connection via regenerated pelvic grafted nerve involved in the control of bladder function. In this study, we attempted to restore surgically ablated bladder function in a rat model using an interposition nerve graft. We chose the genitofemoral nerve because its size approximated that of the pelvic nerve and because loss of the genitofemoral nerve would result in minimal loss of function (loss of cremaster muscle contraction and anesthesia of the upper thigh). Excision of a segment of genitofemoral nerve bilaterally was accomplished in all animals and did not appear to affect the results of bladder function tests. With the exception of week 2 postoperatively, animals with interpositional nerve grafts had better results than those in the ablation group in both bladder function tests and TTC tracing studies. Regrowth of
Kono et al. 677
Surgery Volume 123, Number 6
nerve fibers along the pelvic nerve graft resulted in regeneration of the distal pelvic nerves during the 4-week postoperative period. By week 6, 60% of the grafted animals produced distention-induced rhythmic contractions of the bladder. High-amplitude rhythmic contractions were conducted via the grafted nerve, which had regenerated across a 5 mm gap for 6 weeks, because distention-induced rhythmic contractions were completely eliminated by bilateral section of the grafted nerves. In about 40% to 60% of the sham-operated animals, rapid infusion of saline failed to elicit distention-induced rhythmic contractions, presumably because bladder activation was prevented by an inhibitory sympathetic reflex20,22 or anesthesia. This study showed that nerve grafting can restore bladder function after surgical ablation of the pelvic nerves in rats. Clinical use of our technique may be indicated when intentional surgical ablation of the pelvic nerves is needed to obtain clear surgical margins, such as in surgery for rectal cancer. However, several critical factors must be considered regarding its clinical application in human beings. First, one must decide which nerve is a suitable graft, that is, one must select a nerve that will not result in severe functional disorder if harvested for grafting. Therefore, sensory nerves would be more suitable for grafting than motor nerve. For example, the sural or saphenous nerve could be used. Neither of the two nerves can be removed unless one makes an alternate incision during standard pelvic surgery. The genitofemoral nerve, although it is not a pure sensory nerve, would also be a good candidate because no severe impairment would be observed if it were surgically removed. Furthermore, it could be located easily extending obliquely downward over the psoas muscle during standard pelvic surgery without the need for an alternate incision (Fig. 5, A). In human beings, the genitofemoral nerve is usually much thinner than the pelvic nerve; however, it might be possible to remove the genitofemoral nerve and then reduce it to fragments that can be bundled up to attain the diameter of the pelvic nerve (Fig. 5, B). Second, one must make sure that a correct neurite terminal reconnection is attained. We used a dissection microscope for this purpose in rats. In human beings, it might not be necessary to use a dissection microscope because the nerves are big enough (the hypogastric nerve or pelvic nerve is several millimeters thick). Thus the nerve grafting can be done through epineural anastomosis with glasses for usual microsurgery. Further investigations are necessary to fully under-
Fig. 4. Photomicrograph of TTC-labeled preganglionic neurons in the right lateral intermediate gray of a transverse section (S1) of a nerve-grafted rat at week 6 postoperatively. Bar, 100 µm.
Fig. 5. Scheme illustrates the proposed surgical procedures in human beings. (A) From above downward the ureter rests on the genitofemoral nerve, which is usually seen extending obliquely downward over the psoas muscle. (B) The genitofemoral nerve is removed and reduced to fragments. The fragments are bundled up to attain the diameters of the pelvic nerve.
stand how bladder functions can be restored by pelvic nerve grafting in clinical cases. REFERENCES 1. Aagaard J, Gerstenberg TC, Knudsen JJ. Urodynamic investigation predicts bladder dysfunction at an early stage after abdominoperineal resection of the rectum for cancer. Surgery 1986;99:564-8.
678 Kono et al.
2. Kinn AC, Ohman U. Bladder and sexual function after surgery for rectal cancer. Dis Colon Rectum 1986;29:43-8. 3. Havenga K, DeRuiter MC, Enker WE, Welvaart K. Anatomical basis of autonomic nerve-preserving total mesorectal excision for rectal cancer. Br J Surg 1996;83:384-8. 4. Sugihara K, Moriya Y, Akasu T, Fujita S. Pelvic autonomic nerve preservation for patients with rectal carcinoma: oncologic and functional outcome. Cancer 1996;78:187180. 5. Woodburne RT. The sacral parasympathetic innervation of the colon. Anat Rec 1956;124:67-76. 6. Sunderland S. Advances in diagnosis and treatment of root and peripheral nerve injury. Adv Neurol 1979;22:271-305. 7. Sunderland S. The anatomy and physiology of nerve injury. Muscle Nerve 1990;13:771-84. 8. Sunderland S. The anatomic foundation of peripheral nerve repair techniques. Orthop Clin North Am 1981;12:245-66. 9. Archibald SJ, Krarup C, Shefner J, Li ST, Madison RD. A collagen based nerve guide conduit for peripheral nerve repair: an electrophysiological study of nerve regeneration in rodents and nonhuman primates. J Comp Neurol 1991;306:685-96. 10. Burgers JK, Nelson RJ, Quinlan DM, Walsh PC. Nerve growth factor, nerve grafts and amniotic membrane grafts restore erectile function in rats. J Urol 1991;146:463-8. 11. Dellon AL, Mackinnon SE. An alternative to the classical nerve graft for the management of the short nerve gap. Plast Reconstr Surg 1988;82:849-56. 12. den Dunnen WFA, van der Lei B, Schakenraad JM, et al. Long-term evaluation of nerve regeneration in a biodegradable nerve guide. Microsurgery 1993;14:508-15. 13. Millesi H. Peripheral nerve injuries. Nerve sutures and nerve grafting. Scand J Plast Reconstr Surg 1982;19:25-37. 14. Millesi H. Nerve grafting. Clin Plast Surg 1984;11:105-13. 15. Uzman BG, Villegas GM. Mouse sciatic nerve regeneration through semipermeable tubes. J Neurosci Res 1983;9:32538. 16. Quinlan DM, Nelson RJ, Walsh PC. Cavernous nerve grafts restore erectile function in denervated rats. J Urol 1991;145:380-3. 17. Maggi CA, Santicoli P, Meli A. The nonstop transvesical cystometrogram in urethane anesthetized rats: a simple procedure for quantitative studies on the various phases of urinary bladder voiding cycle. J. Pharmacol Methods 1986;15:157-67. 18. Evinger C, Erichsen JT. Transsynaptic retrograde transport of fragment C of tetanus toxin demonstrated by immunohistochemical localization. Brain Res 1986;380:383-8.
Surgery June 1998 19. Manning KA, Erichsen JT, Evinger C. Retrograde transneuronal transport properties of fragment C of tetanus toxin. Neuroscience 1990;34:251-63. 20. Langworthy OR. Innervation of the pelvic organs of the rat. Invest Urol 1965;2:491-511. 21. Maggi CA, Meli A. Reserpine induced detrusor hyperreflexia: an in vivo model for studying smooth muscle relaxants at urinary bladder level. J Pharmacol Methods 1983;10:79-93. 22. Maggi CA, Santicoli P, Furio M, Meli A. Dual effect of clonidine on micturition reflex in urethane anesthetized rats. J Pharmacol Exp Ther 1985;235:528-36. 23. Heumann R, Lindholm D, Bandtlow C, et al. Differential regulation of mRNA encoding nerve growth factor and its recaptor in rat sciatic nerve during development, degeneration, and regeneration: role of macrophages. Proc Natl Acad Sci U S A 1987;84:8735-9. 24. Taniuchi M, Clerk HB, Schweitzer JB, Johnson EMJ. Expression of nerve growth factor receptors by Schwann cells of axotomized peripheral nerves: ultrastructural location, suppression by axonal contact, and binding properties. J Neurosci 1988;8:664-81. 25. Waris T, Lofstedt T, Partanen S, Smitten KV. Innervation of syngeneic vein grafts in the rat: the regeneration of adrenergic nerves. J Surg Res 1984;37:472-8. 26. Lundborg G, Gelberman RH, Longo FM, Powell HC, Varon S. In vivo regeneration of cut nerves encased in silicone tubes: growth across a six-millimeter gap. J Neuropathol Exp Neurol 1982;41:412-22. 27. Molander H, Olsson Y, Engkvist O, Bowald S, Eriksson I. Regeneration of peripheral nerve through a polyglactin tube. Muscle Nerve 1982;5:54-7. 28. Fine EG, Valentini RF, Bellamkonda R, Aebischer P. Improved nerve regeneration through piezoelectric vinylidenefluoride-trifluoroethylene copolymer guidance channels. Biomaterials 1991;12:775-80. 29. He C, Chen Z, Chen Z. Enhancement of motor nerve regeneration by nerve growth factor. Microsurgery 1992;13:151-4. 30. Laquerriere A, Peulve P, Jin O, et al. Effect of basic fibroblast growth factor and α-melanocytic stimulating hormone on nerve regeneration through a collagen channel. Microsurgery 1994;15:203-10. 31. Lundborg G, Longo FM, Varon S. Nerve regeneration model and trophic factors in vivo. Brain Res 1982;232:15761. 32. Hancock MB. Preganglionic neurons in the sacral spinal cord of the rat: an HRP study. Neurosci Lett 1979;11:1-5.
Background. Autonomic nerve preservation techniques for use during surgery for rectal cancer have improved. Nevertheless, in some patients pelvic nerves must be sacrificed to excise all tumor. For these patients, nerve reconstruction at the time of operation by using nerve grafting would be useful. A rat model of this type of nerve reconstruction is described. Methods. Animals were divided into three groups. In the sham control group, a pelvic exploration was conducted without division of the pelvic nerves. In the nerve ablation group, both pelvic nerves were excised segmentally. In the graft group, both pelvic nerves were excised and genitofemoral nerves were interposed bilaterally. At 2-week intervals postoperatively, animals from each group underwent cystometry under urethane anesthesia and neuronal tracing using fragment C of tetanus toxin for demonstration of axonal transport via regenerated nerves. Results. At 6 weeks postoperatively, 60% of the grafted animals produced rhythmic contractions of the bladder. In neuronal tracing studies at weeks 4 and 6, respectively, 40% and 100% of the nerve-grafted rats had labeled neurons in the sacral parasympathetic nucleus. Conclusions. These findings suggest that pelvic nerve grafting in rats can successfully restore bladder function after surgical injury. Clinical use of pelvic nerve grafting may be indicated for patients whose pelvic nerves must be sacrificed to excise all tumor.(Surgery 1998;123:672-8.) From the Second Department of Surgery and the Department of Urology, Asahikawa Medical College, Hokkaido, Japan
SURGICAL TREATMENT FOR RECTAL CANCER has profound effects on bladder function resulting from autonomic denervation.1,2 Recent studies have precisely localized the nerves that control bladder function.3-5 With this knowledge it has been possible to preserve the autonomic nerves controlling bladder function in many patients undergoing radical resection for rectal cancer. However, in some patients with advanced rectal cancer, autonomic nerve preservation cannot be performed if adequate surgical margins are to be obtained. At the present time, no methods are available for restoring bladder function in these patients. Therefore, methods to restore neural control of bladder function without compromising rectal cancer control must be devised. The feasibility of nerve regeneration in denervated tissues of the peripheral nervous system has been carefully studied.6-8 In addiSupported by research grant (C)(2)07671350 from the Japanese Ministry of Education, Science, and Culture. Accepted for publication Nov. 5, 1997. Reprint requests: Toru Kono, MD, PhD, The Second Department of Surgery, Asahikawa Medical College, 4-5 Nishikagura, Asahikawa, Hokkaido 078, Japan. Copyright © 1998 by Mosby, Inc. 0039-6060/98/$5.00 + 0 11/56/87454
672 SURGERY
tion, nerve grafting is a well-established technique in plastic surgery for repair of damaged nerves.9-16 However, the use of such techniques for autonomic nerves controlling bladder function has not been studied. Methods enabling direction of regenerating autonomic nerve would be valuable for patients who require intentional surgical pelvic nerve denervation to ensure clear surgical margins. We demonstrate here that nerve grafting with transposed genitofemoral nerve can restore bladder function in rats that had previously undergone surgical nerve ablation. This study demonstrates the feasibility of restoration of bladder function after advanced cancer surgery by interpositional nerve grafting of the pelvic nerves. MATERIAL AND METHODS Forty-five Wistar rats were divided into three groups of 15 animals each: (1) sham controls, (2) pelvic nerve ablation, and (3) pelvic nerve grafting. Postoperatively, each 2 weeks for 6 weeks five animals from each group underwent postoperative bladder function tests17 and examination for restoration of axonal transport between the regenerated sciatic nerve and the spinal cord, principally its lumbosacral portion, using fragment C of tetanus toxin (TTC) retrograde tracing methods.18,19
Surgery Volume 123, Number 6
Kono et al. 673
Fig. 1. Scheme illustrates surgical procedures in rats. (A) The pelvic ganglion receives input from the pelvic nerve, which is the final preganglionic parasympathetic pathway controlling bladder function in the rat. (B) In the nerve ablation group, a 5 mm segment was excised from both pelvic nerves. (C) In the graft group, a 5 mm segment was excised from both pelvic nerves and replaced with interpositional grafts of genitofemoral nerves bilaterally. A nerve graft sewn in place restored continuity of the pelvic nerve sheath.
Surgical procedures. Seventy-day-old Wistar rats (250 to 270 gm) were used. All animals were anesthetized with intraperitoneal sodium pentobarbital (25 mg/kg) and underwent surgical exploration to expose the innervation of the pelvic viscera and lower genitourinary tract.20 A midline incision was made from umbilicus to pubis. With the aid of a Nikon dissecting microscope (magnification ×8), the hypogastric nerve, pelvic nerve, major pelvic ganglion, and the nerve fibers to the lower genitourinary tract were identified20 (Fig. 1). The pelvic nerves were identified and exposed bilaterally. In the sham control group, the pelvic nerves were identified but not divided at the time of operation (Fig. 1, A). In the pelvic nerve ablation group, a 5 mm segment of pelvic nerve just distal to the major pelvic ganglion was surgically excised bilaterally (Fig. 1, B). Care was taken to excise all branches of the pelvic nerve trunk, because unsuccessful denervation has been reported when the lateral branches have been spared. In the pelvic nerve graft group, a 5 mm segment of pelvic nerve and all accessory branches were excised bilaterally, and an isograft of genitofemoral nerve was interposed bilaterally (Fig. 1, C). The direction of the graft was reversed so that the bifurcating distal end of the genitofemoral nerve (dividing into its genital and femoral branches) was placed proximally. This method of positioning allows regenerating nerve fibers to converge distally.16 An epineural
anastomosis (proximal and distal) was then performed with 10-0 nylon sutures using microsurgical technique. Bladder function tests. Cystometric bladder function studies in the urethane-anesthetized rat have been well described.17,21,22 Every 2 weeks postoperatively for 6 weeks, five animals from each of the three experimental groups underwent bladder function tests. All bladder function tests were performed by transvesical recording of intravesical pressure. Rats were anesthetized with urethane (1.5 gm/kg, subcutaneously). Through a midline incision of the abdomen, the urinary bladder was exposed and a Teflon catheter (Angiocath, 18gauge; Deseot Medical Inc.) was implanted into the bladder through the dome and secured in place by means of a silk ligature. The bladder catheter was connected via a T-tube to a pressure transducer and a microinjection pump. The urethra was ligated between the prostate and the bladder using a silk thread suture. Warm saline-soaked cotton wool swabs were laid around the exteriorized organ to maintain its temperature and keep it moist. After a 15-minute equilibration period at zero volume, the bladder was rapidly filled with 0.5 ml of warm saline (37° C) to increase and maintain intravesical pressure. This maneuver produced neurogenic rhythmic contractions. Intravesical pressure was recorded continuously on a San-ei polygraph (Model 363; Nippon Koden San-ei) and displayed on a poly-
674 Kono et al.
Fig. 2. A representative tracing illustrates results of bladder function testing for a sham-operated rat (Sham) and a rat that had undergone bilateral removal of the pelvic nerves (Bilateral Denervation). IVP, intravesical pressure.
graph. At the beginning of cystometric measurements, the bladder was emptied through the bladder catheter. In the pelvic nerve graft group, cystometry was performed before and after the grafted pelvic nerve was removed. The rhythmic contractions required 5 to 10 minutes to reach steady state after saline loading. The frequency, duration, and amplitude of rhythmic contractions were determined at this time to evaluate the effects of each preparation. The number, width, and height of contractions were determined during a 4-minute observation period. TTC tracing studies. After bladder function tests, animals underwent TTC tracing tests. TTC (10 µl) was slowly injected into the ventral bladder detrusor muscle using a Hamilton microliter syringe fitted with a 27-gauge needle, and the incision was closed. The TTC (Behring DiagnosticsCalbiochem; molecular weight, 47,000) had been concentrated to approximately 15% (w/v) by ultrafiltration with a Centricon tube (Amicon; 30,000 molecular weight cutoff membrane). All injections were made into animals anesthetized with urethane. Optimal survival times for transsynaptic labeling were determined for normal rats in which TTC had been injected into the wall of the bladder. Transsynaptic labeling was well developed by 3 days
Surgery June 1998 after injection. Therefore, after a transport time of 72 hours for TTC, each animal was deeply anesthetized. Injection of 1000 units of heparin preceded transcardial perfusion with 6% dextran followed by 4% paraformaldehyde in 0.1 mol/L phosphate buffer (PB) at pH 7.4. After perfusion, the lower spinal cord was removed, postfixed in buffered 4% paraformaldehyde at 4° C for 3 hours, and placed in 30% sucrose in PB at 4° C for 24 hours. The spinal cord was subsequently frozen and cut frontally in serial order at 10 µm thickness in a cryostat, serially mounted on gelatinized slides, and heat-dried on a slide warmer at 40° to 50° C. Immunohistochemistry for TTC. TTC was localized immunohistochemically. The sections were treated with endogenous peroxidase (0.2% H2O2 in distilled water for 10 minutes) and incubated overnight at 4° C in a monoclonal antibody that had been raised in mouse against TTC. The antibody (ascites fluid) was diluted 1:500 in 0.3% Triton-X and PB (TX/PB). The sections were washed in PB, incubated at room temperature for 2 hours in biotinylated anti-mouse immunoglobulin G diluted 1:200 in TX/PB, washed again in PB, and then incubated for 1 hour in avidin-biotinhorseradish peroxidase complex (1:50 in TX/PB) using standard procedures and reagents from Vector Laboratories. In all cases, a 5-minute incubation in 3% cobalt chloride in PB was included to intensify the reaction. The tissue was then reacted with diaminobenzidine (55 mg/1,000 ml PB) with 0.03%H2O2. Sections were dried and incubated briefly in a 0.05% osmium tetroxide solution before dehydration and coverslipping. In control experiments, tissue unexposed to TTC was processed normally, and tissue containing TTC was processed without the primary antibody. Both sets of control experiments yielded no labeling. Statistical analyses. Results are given as mean values ± standard error of the mean. Statistical analysis was performed using Student’s t test for paired or unpaired data as appropriate. Findings were considered significant when p < 0.05. RESULTS Bladder function tests. At 2-week intervals postoperatively, five animals from each group underwent bladder function testing (Table I). Rapid distention of the urinary bladder by infusion of saline in sham animals activated a series of high-amplitude rhythmic contractions that were eliminated by bilateral section of the pelvic nerves (Fig. 2). At 2 and 4 weeks, distention of the urinary bladder by infusion of saline in the grafted animals failed to elicit neurogenic rhythmic contractions
Kono et al. 675
Surgery Volume 123, Number 6
Table I. Percentages of animals producing distention-induced rhythmic contractions of the bladder Treatment groups Sham Ablation Graft
2 wk
4 wk
6 wk
60 (3/5) 0 (0/5) 0 (0/5)
40 (2/5) 0 (0/5) 0 (0/5)
60 (3/5) 0 (0/5) 60 (3/5)
Sham, control group (sham procedure); ablation, animals with bilateral excision pelvic nerves; graft, animals with pelvic nerve graft.
Table II. Comparison of amplitude and duration of distention-induced rhythmic bladder contraction at week 6 postoperatively Treatment groups Sham Graft
Frequency (count/min)
Amplitude (cm H2O)
1.3 ± 0.1* 0.7 ± 0.1*
37.0 ± 0.2† 26.6 ± 0.7†
Duration (sec) 27.7 ± 0.2‡ 25.4 ± 0.7‡
*p < 0.05. †p
< 0.05.
‡No
significant difference.
Table III. Percentages of animals with TTC-labeled neurons in the spinal cord Treatment groups Sham Ablation Graft
2 wk
4 wk
6 wk
100 (5/5) 0 (0/5) 0 (0/5)
100 (5/5) 0 (0/5) 40 (2/5)
100 (5/5) 0 (0/5) 100 (5/5)
Sham, control group (sham procedure); ablation, animals with bilateral excision pelvic nerves; graft, animals with pelvic nerve graft.
(Table I). At 6 weeks, 60% of the grafted animals produced distention-induced rhythmic contractions of the bladder (Table I). The amplitude and frequency of these contractions were significantly (p < 0.05) lower than those in control animals, but the duration was not (Table II). These contractions in the grafted animals disappeared after bilateral section of the grafted pelvic nerves (Fig. 3). None of the animals in the ablation group produced high-amplitude rhythmic contractions during the 6-week period of observation (Table I). These findings indicated that at 6 weeks the grafted animals had approached the sham-operated group in their ability to produce bladder contractions, whereas animals in the nerve ablation group were still unable to produce such contractions. TTC tracing studies. Results are shown in Table III and Fig. 4. At 2 weeks, no labeled cells were found in the spinal cord in either the ablation or graft group. In the sham group, many preganglionic neurons labeled by TTC were found in the sacral parasympathetic nucleus (SPN) in the L6-S1 spinal cord; the majority of labeled cells were located in S1. In transverse sections, the TTC-labeled cells were most frequently observed in tight clusters in the lateral intermediate gray (Fig. 4). At week 4, 40% of the grafted animals had labeled neurons in
the SPN, but the number of labeled neurons was much lower than in control animals. At week 6, labeled neurons were observed in the SPN in 100% of the grafted animals, and the number of labeled neurons was greater than at week 4. No staining of lumbosacral parasympathetic neurons was observed in the nerve-ablated animals. These findings demonstrated that regrowth of nerve fibers along the pelvic nerve graft had resulted in regeneration of the distal pelvic nerves during the 4- to 6week postoperative period. DISCUSSION Clinical interest in bladder function after surgical resection (e.g., abdominoperineal resection, low anterior resection) for advanced rectal cancer has resulted from dissatisfaction with what had been accepted as the necessary loss of such function after this operation. It is now possible to identify the location of the pelvic nerves intraoperatively and, in certain patients, to avoid injuring them, thus preserving autonomic nerve control of bladder function. In some patients, however, it is necessary to excise both pelvic nerves to obtain clear surgical margins. Unfortunately, the chance of preserving natural bladder function in these patients has been very small. Patients undergoing
676 Kono et al.
Fig. 3. A representative tracing illustrates results of bladder function testing at week 6 for a rat with pelvic nerve graft (Graft) and a rat that had undergone bilateral removal of the grafted nerves (Bilateral Denervation). IVP, intravesical pressure.
partial excision of the pelvic nerves for malignant disease also face a serious risk of urinary dysfunction. No work has focused on procedures for restoring natural bladder function without compromising focal control of rectal cancer and elimination of recurrence in such patients. It has been demonstrated in animal models that autonomic peripheral nerves can regenerate and reestablish terminal contacts. The use of direct anastomosis or autograft in peripheral nerve repair leads to functional recovery. Nerve grafting is a well-established technique in plastic surgery for repair of damaged motor and sensory nerves9,10,13,14,16 A nerve graft provides a conduit down which regenerating nerve fibers will be directed to meet with the distal end of the transected nerve. Wallerian degeneration in the distal nerve segment (as well as the graft segment itself) occurs once the nerve loses continuity with its proximal segment and cell body. This leaves only the nerve sheath intact. Regenerating fibers may then grow down this sheath directed toward the distal nerve while protected from fibrosis and neuroma formation.13 Degenerating nerve fibers in the distal nerve and graft also may provide nerve growth factor, which will both stimulate nerve fiber growth and direct it toward the distal end of the
Surgery June 1998 nerve.23,24 Nerve graft techniques used by other investigators in rats have demonstrated that cavernous nerve grafting in rats can restore potency after surgical injury.16 To our knowledge, no studies of restoration of bladder function using nerve grafts have been reported. Venous grafts have been shown to facilitate autonomic nerve regeneration.25 In recent years, many biodurable and biodegradable nerve guide conduits manufactured from silicone, polyvinylidone fluoride, propylene, polyglycolic acid, or collagen have been tested.9,11,12,15,26-28 Another approach to improving the success of nerve repair procedures is the search for trophic factors.10,29-31 These factors promote long nerve gap regeneration better than grafting alone. The next step in our work will be to find nerve guide conduits and trophic factors appropriate for restoration of bladder function that can be used in place of nerve grafting. TTC is a nontoxic molecule that undergoes robust retrograde transneuronal transport from the periphery and can be localized using a monoclonal antibody.18,19 The presence of labeling in SPN, one of the portions of the central nervous system controlling detrusor contraction,32 demonstrates that transneuronal transport via regenerated grafted nerve occurs. After injection into the detrusor muscle of the bladder, TTC is taken up by the axon terminals of postganglionic neurons in the major pelvic ganglion and transported in a retrograde fashion to their somas. TTC then moves across synapses to presynaptic elements and passes the pelvic grafted nerve in a retrograde fashion back to the somas of preganglionic neurons in the SPN in the spinal cord. Therefore, retrograde transneuronal transport from the bladder detrusor muscle to the SPN can only be explained by the presence of neuroanatomic connection via regenerated pelvic grafted nerve involved in the control of bladder function. In this study, we attempted to restore surgically ablated bladder function in a rat model using an interposition nerve graft. We chose the genitofemoral nerve because its size approximated that of the pelvic nerve and because loss of the genitofemoral nerve would result in minimal loss of function (loss of cremaster muscle contraction and anesthesia of the upper thigh). Excision of a segment of genitofemoral nerve bilaterally was accomplished in all animals and did not appear to affect the results of bladder function tests. With the exception of week 2 postoperatively, animals with interpositional nerve grafts had better results than those in the ablation group in both bladder function tests and TTC tracing studies. Regrowth of
Kono et al. 677
Surgery Volume 123, Number 6
nerve fibers along the pelvic nerve graft resulted in regeneration of the distal pelvic nerves during the 4-week postoperative period. By week 6, 60% of the grafted animals produced distention-induced rhythmic contractions of the bladder. High-amplitude rhythmic contractions were conducted via the grafted nerve, which had regenerated across a 5 mm gap for 6 weeks, because distention-induced rhythmic contractions were completely eliminated by bilateral section of the grafted nerves. In about 40% to 60% of the sham-operated animals, rapid infusion of saline failed to elicit distention-induced rhythmic contractions, presumably because bladder activation was prevented by an inhibitory sympathetic reflex20,22 or anesthesia. This study showed that nerve grafting can restore bladder function after surgical ablation of the pelvic nerves in rats. Clinical use of our technique may be indicated when intentional surgical ablation of the pelvic nerves is needed to obtain clear surgical margins, such as in surgery for rectal cancer. However, several critical factors must be considered regarding its clinical application in human beings. First, one must decide which nerve is a suitable graft, that is, one must select a nerve that will not result in severe functional disorder if harvested for grafting. Therefore, sensory nerves would be more suitable for grafting than motor nerve. For example, the sural or saphenous nerve could be used. Neither of the two nerves can be removed unless one makes an alternate incision during standard pelvic surgery. The genitofemoral nerve, although it is not a pure sensory nerve, would also be a good candidate because no severe impairment would be observed if it were surgically removed. Furthermore, it could be located easily extending obliquely downward over the psoas muscle during standard pelvic surgery without the need for an alternate incision (Fig. 5, A). In human beings, the genitofemoral nerve is usually much thinner than the pelvic nerve; however, it might be possible to remove the genitofemoral nerve and then reduce it to fragments that can be bundled up to attain the diameter of the pelvic nerve (Fig. 5, B). Second, one must make sure that a correct neurite terminal reconnection is attained. We used a dissection microscope for this purpose in rats. In human beings, it might not be necessary to use a dissection microscope because the nerves are big enough (the hypogastric nerve or pelvic nerve is several millimeters thick). Thus the nerve grafting can be done through epineural anastomosis with glasses for usual microsurgery. Further investigations are necessary to fully under-
Fig. 4. Photomicrograph of TTC-labeled preganglionic neurons in the right lateral intermediate gray of a transverse section (S1) of a nerve-grafted rat at week 6 postoperatively. Bar, 100 µm.
Fig. 5. Scheme illustrates the proposed surgical procedures in human beings. (A) From above downward the ureter rests on the genitofemoral nerve, which is usually seen extending obliquely downward over the psoas muscle. (B) The genitofemoral nerve is removed and reduced to fragments. The fragments are bundled up to attain the diameters of the pelvic nerve.
stand how bladder functions can be restored by pelvic nerve grafting in clinical cases. REFERENCES 1. Aagaard J, Gerstenberg TC, Knudsen JJ. Urodynamic investigation predicts bladder dysfunction at an early stage after abdominoperineal resection of the rectum for cancer. Surgery 1986;99:564-8.
678 Kono et al.
2. Kinn AC, Ohman U. Bladder and sexual function after surgery for rectal cancer. Dis Colon Rectum 1986;29:43-8. 3. Havenga K, DeRuiter MC, Enker WE, Welvaart K. Anatomical basis of autonomic nerve-preserving total mesorectal excision for rectal cancer. Br J Surg 1996;83:384-8. 4. Sugihara K, Moriya Y, Akasu T, Fujita S. Pelvic autonomic nerve preservation for patients with rectal carcinoma: oncologic and functional outcome. Cancer 1996;78:187180. 5. Woodburne RT. The sacral parasympathetic innervation of the colon. Anat Rec 1956;124:67-76. 6. Sunderland S. Advances in diagnosis and treatment of root and peripheral nerve injury. Adv Neurol 1979;22:271-305. 7. Sunderland S. The anatomy and physiology of nerve injury. Muscle Nerve 1990;13:771-84. 8. Sunderland S. The anatomic foundation of peripheral nerve repair techniques. Orthop Clin North Am 1981;12:245-66. 9. Archibald SJ, Krarup C, Shefner J, Li ST, Madison RD. A collagen based nerve guide conduit for peripheral nerve repair: an electrophysiological study of nerve regeneration in rodents and nonhuman primates. J Comp Neurol 1991;306:685-96. 10. Burgers JK, Nelson RJ, Quinlan DM, Walsh PC. Nerve growth factor, nerve grafts and amniotic membrane grafts restore erectile function in rats. J Urol 1991;146:463-8. 11. Dellon AL, Mackinnon SE. An alternative to the classical nerve graft for the management of the short nerve gap. Plast Reconstr Surg 1988;82:849-56. 12. den Dunnen WFA, van der Lei B, Schakenraad JM, et al. Long-term evaluation of nerve regeneration in a biodegradable nerve guide. Microsurgery 1993;14:508-15. 13. Millesi H. Peripheral nerve injuries. Nerve sutures and nerve grafting. Scand J Plast Reconstr Surg 1982;19:25-37. 14. Millesi H. Nerve grafting. Clin Plast Surg 1984;11:105-13. 15. Uzman BG, Villegas GM. Mouse sciatic nerve regeneration through semipermeable tubes. J Neurosci Res 1983;9:32538. 16. Quinlan DM, Nelson RJ, Walsh PC. Cavernous nerve grafts restore erectile function in denervated rats. J Urol 1991;145:380-3. 17. Maggi CA, Santicoli P, Meli A. The nonstop transvesical cystometrogram in urethane anesthetized rats: a simple procedure for quantitative studies on the various phases of urinary bladder voiding cycle. J. Pharmacol Methods 1986;15:157-67. 18. Evinger C, Erichsen JT. Transsynaptic retrograde transport of fragment C of tetanus toxin demonstrated by immunohistochemical localization. Brain Res 1986;380:383-8.
Surgery June 1998 19. Manning KA, Erichsen JT, Evinger C. Retrograde transneuronal transport properties of fragment C of tetanus toxin. Neuroscience 1990;34:251-63. 20. Langworthy OR. Innervation of the pelvic organs of the rat. Invest Urol 1965;2:491-511. 21. Maggi CA, Meli A. Reserpine induced detrusor hyperreflexia: an in vivo model for studying smooth muscle relaxants at urinary bladder level. J Pharmacol Methods 1983;10:79-93. 22. Maggi CA, Santicoli P, Furio M, Meli A. Dual effect of clonidine on micturition reflex in urethane anesthetized rats. J Pharmacol Exp Ther 1985;235:528-36. 23. Heumann R, Lindholm D, Bandtlow C, et al. Differential regulation of mRNA encoding nerve growth factor and its recaptor in rat sciatic nerve during development, degeneration, and regeneration: role of macrophages. Proc Natl Acad Sci U S A 1987;84:8735-9. 24. Taniuchi M, Clerk HB, Schweitzer JB, Johnson EMJ. Expression of nerve growth factor receptors by Schwann cells of axotomized peripheral nerves: ultrastructural location, suppression by axonal contact, and binding properties. J Neurosci 1988;8:664-81. 25. Waris T, Lofstedt T, Partanen S, Smitten KV. Innervation of syngeneic vein grafts in the rat: the regeneration of adrenergic nerves. J Surg Res 1984;37:472-8. 26. Lundborg G, Gelberman RH, Longo FM, Powell HC, Varon S. In vivo regeneration of cut nerves encased in silicone tubes: growth across a six-millimeter gap. J Neuropathol Exp Neurol 1982;41:412-22. 27. Molander H, Olsson Y, Engkvist O, Bowald S, Eriksson I. Regeneration of peripheral nerve through a polyglactin tube. Muscle Nerve 1982;5:54-7. 28. Fine EG, Valentini RF, Bellamkonda R, Aebischer P. Improved nerve regeneration through piezoelectric vinylidenefluoride-trifluoroethylene copolymer guidance channels. Biomaterials 1991;12:775-80. 29. He C, Chen Z, Chen Z. Enhancement of motor nerve regeneration by nerve growth factor. Microsurgery 1992;13:151-4. 30. Laquerriere A, Peulve P, Jin O, et al. Effect of basic fibroblast growth factor and α-melanocytic stimulating hormone on nerve regeneration through a collagen channel. Microsurgery 1994;15:203-10. 31. Lundborg G, Longo FM, Varon S. Nerve regeneration model and trophic factors in vivo. Brain Res 1982;232:15761. 32. Hancock MB. Preganglionic neurons in the sacral spinal cord of the rat: an HRP study. Neurosci Lett 1979;11:1-5.